There have been rudimentary attempts to develop machines that respond to free hand indicators of control commands. Certain attempts have addressed localized hand control regions in the view of one or more cameras using image structure methods or statistical learning for user interface (UI) control. Some are as simple as “on/off” control by waving a hand before a sensor of some type (e.g. photosensor or infrared (IR) sensor). Other methods involve defining a space in a region dictated by the camera system setup where only hand gestures within the defined region are accepted by the interaction system. However, these systems dictate a three-dimensional control area and the user needs to gesture within the control area, which may not necessarily be convenient for the user. These systems may also be challenged in regards to efficiently defining and recognizing gestures which fit a user's preference, and avoiding accidental gestures in gesture interactions, particularly in conjunction with an automotive UI.